We enrolled 212 patients into five groups: including those with disseminated nontuberculous mycobacterial disease, severe opportunistic infections, pulmonary tuberculosis, disseminated tuberculosis, and normal blood bank controls. We identified anti-interferon gamma autoantibodies in approximately 90% of those with severe opportunistic infections. All these patients were HIV uninfected and had no other recognized cause of immune dysfunction. Interestingly, one patient with cryptococcal meningitis had anti-GM-CSF autoantibodies. Therefore, we have shown in a large cohort of patients in Thailand and Taiwan with severe opportunistic infections, including nontuberculous mycobacteria, that autoantibodies to interferon gamma account for the defect and reproduce a state of severe immunodeficiency. As a result of this project and in order to extend these diagnostic opportunities to the field, we have developed a simple screening assay for anti-interferon gamma autoantibodies that allows us to identify, follow, and titer anti-interferon gamma activity. The recognition of anti-interferon gamma autoantibodies as the cause of severe opportunistic infections has led to the use of rituximab for control of their antibodies and therefore their infections. These conditions overlap considerably with advanced AIDS, except for the absence of pneumocystis. We are now engaged in characterizing the epitope or epitopes that are being recognized by these autoantibodies. We have also identified autoantibodies to GM-CSF in Cryptococcus gattii infection and extra pulmonary Nocardia infection.